BigHugger
GH Repository · AgentR1

Agent-R1

Agent-R1: Training Powerful LLM Agents with End-to-End Reinforcement Learning

stars
1,669
30-day movement
+83/day
Related entries
8
Connections
1
pythonagent-appPythonllmagentic-rlagent

Agent-R1 is a Python project for training LLM agents with end-to-end reinforcement learning. Per its title, it focuses on the agentic-RL training approach rather than on serving or deploying trained agents.

It gives you a codebase dedicated to training LLM agents via end-to-end reinforcement learning, which you would otherwise have to build yourself.

Use it to

  • Train LLM agents with end-to-end reinforcement learning
  • Experiment with agentic-RL methods
  • Study the repository's training pipeline structure
  • Build custom agent training setups on its codebase

For Researchers and engineers training RL-based LLM agents

Role
agent-app
Language
Python
Licence
MIT
Forks
117
Open issues
34
Last push
2026-09-14
topicsagentsreinforcement-learningllmtrainingpythonresearch